I am a postdoc at the University of Chicago Data Science Institute, working with Mina Lee. My research interests lie at the intersection of Human-Computer Interaction, Social Computing, and Responsible AI. I design and evaluate human-centered technology to support users' learning and knowledge work, focusing primarily on online education and AI-assisted writing contexts. I approach this goal from two angles: (1) designing interactive systems that scaffold cognitive processes in sensemaking and content creation, and (2) fostering responsible use of technology.

I previously earned a PhD in Computer Science from UC Davis, advised by Hao-Chuan Wang, and dual BSc degrees in Mathematics and Computer Science from the University of Minnesota, Twin Cities.

News

  • Aug 2026 Our paper about educational chatbot personas was accepted to the EMNLP NLP4PI Workshop.
  • Jun 2026 Attended FAccT in Montreal and presented a paper about readers' and writers' perceived necessity of AI disclosure.
  • Jun 2026 Attended ICLS in Irvine and presented our poster about LLM-generated conversation initiations in video-based learning.
  • Jun 2026 Our work about social norms in knowledge-based online communities was published in the Computer Supported Cooperative Work (CSCW) journal.
  • Oct 2025 Named the DSI ZhengTong Postdoctoral Fellow for 2025-2026 in recognition of outstanding research.

Selected Publications

See the full publication list on Google Scholar.

Interactive Systems for Online Education

EduLive: Re-Creating Cues for Instructor-Learners Interaction in Educational Live Streams with Learners' Transcript-Based Annotations

Jingchao Fang, Jeongeon Park, Juho Kim, Hao-Chuan Wang

CSCW 2024 · Proceedings of the ACM on Human-Computer Interaction

TL;DR: EduLive is a live-streaming support system that turns learners' transcript-based annotations into interactivity cues for teachers.

Understanding the Effects of Structured Note-taking Systems for Video-based Learners in Individual and Social Learning Contexts

Jingchao Fang, Yanhao Wang, Chi-Lan Yang, Ching Liu, Hao-Chuan Wang

GROUP 2022 · Proceedings of the ACM on Human-Computer Interaction

TL;DR: NoteCoStruct is a digital note-taking and note-sharing tool that turns fragmented notes into shareable learning traces.

How Persona and Message Type in Conversation Initiation Influence Learners’ Experience with Educational Chatbots

Jingchao Fang, Qingxiaoyang Zhu, Hao-Chuan Wang

EMNLP 2026 NLP4PI Workshop · 5th Workshop on NLP for Positive Impact.

TL;DR: Using EduPrompt, we find that learning companion chatbot's personas and message styles influence learners' perceptions and reactions in video-based learning.

Responsible AI Use in Knowledge Work

What Influences Readers' and Writers' Perceived Necessity of AI Disclosure?

Jingchao Fang, Victoria Xiaohan Wen, Mina Lee

FAccT 2026 · ACM Conference on Fairness, Accountability, and Transparency

TL;DR: We studied how perspective (reader or writer of the text), purpose (the goal of reading or writing), and procedural factors (how AI was used) influences perceived necessity of AI disclosure in AI-assisted writing.

Fostering Mindful Human-LLM Interaction with Design Friction

Jingchao Fang et al. Paper title altered for anonymization

Under Review

TL;DR: We introduce voluntary friction to foster mindful LLM use in AI-assisted writing.

On LLM Wizards: Identifying Large Language Models' Behaviors for Wizard of Oz Experiments

Jingchao Fang, Nikos Arechiga, Keiichi Namaoshi, Nayeli Bravo, Candice Hogan, David A. Shamma

IVA 2024 · ACM International Conference on Intelligent Virtual Agents

TL;DR: We introduce an experiment lifecycle that guides researchers to responsibly integrate LLMs into Wizard of Oz experiments.

Online Community

Understanding Social Influence in Collective Product Ratings Using Behavioral and Cognitive Metrics

Fu-Yin Cherng, Jingchao Fang, Yinhao Jiang, Xin Chen, Taejun Choi, Hao-Chuan Wang

CHI 2022 · ACM Conference on Human Factors in Computing Systems

TL;DR: We conducted a lab experiment to investigate how products' ratings and reviews influence users' preferences and cognitive responses assessed by their EEG.

Understanding Discussions Around Norms in Heavily Moderated Knowledge-Based Online Communities: A Case Study on Meta Stack Overflow

Jingchao Fang, Jia-Wei Liang, Hao-Chuan Wang

Computer Supported Cooperative Work: The Journal of Collaborative Computing and Work Practices

TL;DR: We established a taxonomy of "what's and hows" of norm-related discussions in a heavily moderated, knowledge-based online community.